Sound Change Quickie Thread

[Nortaneous]

KQ used to claim it happened in Danish.

[Pogostick Man]

I'm sure somebody has asked this before, but I've had no luck finding it. It's famously rare for palatal or palatalized velar stops to shift to plain velars; I was just curious if anybody knew of any case where this is known or believed to have happened unconditionally.

This may have happened in Egyptian Arabic, which has [g] where the standard has [dʒ]. It depends on what you consider the ancestor to the sound was (if it was *ɟ or *gʲ, or just *g).

[----]

Depending on what you think Proto-Austronesian *j is, that happened in Pangasinan and several other languages of the Philippines: *j > g, e.g. *qalejaw > Pang. agew; *bajaq > Ilokano baga

[vokzhen]

Depending on what you think Proto-Austronesian *j is, that happened in Pangasinan and several other languages of the Philippines: *j > g, e.g. *qalejaw > Pang. agew; *bajaq > Ilokano baga

Given *j is reflected almost entirely as coronal, it seems likely the route was more like Pashto retroflex>velar or Irish dental>velar (or Spanish/Slavic postalveolar/velar, though those particular examples are both voiceless).

[----]

None of the experts reconstruct it as dental or retroflex, and it's reflected as various palatals in oceanic, so I find that somewhat more tenuous. than the standard assumed historical value.

[vokzhen]

Unless Wikipedia is wrong (or wildly unrepresentative), *j looks to be dental/alveolar in most Formosan languages and coronal in all except maybe a couple isolated cases, which accounts for somewhere between 3/4ths and 9/10ths Austronesian branches. If Oceanic has a palatal (and some Philippine a velar), it seems very likely that's a secondary development. Though I don't know the details of why they construct a palatal, I'm just looking at Wikipedia's table of correspondences, so maybe I'm talking out my ass. (Though taking a closer look, it looks like Blust does reconstruct it as a coronal *z, and unless I'm misreading Ross doesn't accept that *j/*z is even distinct from *d until some time after Malayo-Polynesian split off.) A good way to screw things up is to read them backwards.

[StrangerCoug]

Let's say you're starting out with something like /a e ɛ i ɔ o u p b m t d n t͡ʃ d͡ʒ ɲ k ɡ ŋ ʔ h s z ʃ ʒ r/ for your inventory and you allow (C)V(C) syllables with the restriction that the coda consonant, if it's there, can't be any of /b d d͡ʒ ɡ z ʒ/ (at least not until you apply allophonic voicing). /b d d͡ʒ ɡ z ʒ/ happen to be the very sounds you lose to a tone split; syllables that once started with those consonants are now pronounced with a low tone while the other syllables are pronounced with a high tone. If you want the tone system to be more complex, what are some developments it can take from there?

[vokzhen]

Losing coda fricatives (likely /h/ in this case) lowers tone or causes a falling tone. Losing coda glottal stops can be either a lower/falling tone or a higher/rising tone (not sure there's any accepted evidence for the difference, but I think I've heard stiff versus creaky voice, pure glottal stop versus creaky voice, creaky voice versus harsh voice, etc). Hiatus with a tone on each can merge into a single vowel + a contour tone. Not entirely sure it's attested but diphthongs might monophthongize with a high-falling or low-falling tone (or high-rising or low-rising, if they're rising diphthongs); similarly, it's attested that a sequence of vowel + glide + vowel can monophthongize with contour tone, and I'd imagine you can get register tones in that way too (VhV > breathy + tone, VʔV > glottalized + tone).

[Nortaneous]

sagart reconstructs *j as a palatal nasal w/ similar developments to chinese

[----]

Yeah but Sagart thinks there was such a thing as "Sino-Austronesian" and PA already has a palatal nasal, so

[Nortaneous]

sagart thinks *ñ was a secondary development arising from palatalization of *n *N (which merged in kanakanavu and PMP) in the environment _iV. *ñ in other Formosan languages sometimes behaves as *n and sometimes as *N, which makes sense if they were in fact *n and *N

[احمکي ارش-ھجن]

Does anyone know what kinds of consonant clusters aside from sC clusters would merit a prosthesis (word-initial epenthesis) of /i/? Voiced-stop+liquid perhaps?

[----]

anything beginning with /r/ or other rhotics

[Porphyrogenitos]

With the case of /r/ it doesn't even have to be a cluster to trigger prosthesis - this happened in some Turkic and Balkan Romance languages, as exhibited in the names of the Aromanian and Urum languages.

[احمکي ارش-ھجن]

With the case of /r/ it doesn't even have to be a cluster to trigger prosthesis - this happened in some Turkic and Balkan Romance languages, as exhibited in the names of the Aromanian and Urum languages.

I don't want/like onset voiced-stop+liquid clusters but I don't want epenthesis to break them up.

Perhaps there was epenthesis and then metathesis occurs to make it look like prosthesis?

[sangi39]

I can't remember exactly where I was going with this (possibly some sort of "initial mutation" scenario after reading something on Facebook), but does > [β ð ɣ] adjacent to [r] (and possibly [l] and maybe [z]) seem plausible, with remaining as such in all other positions?

I might be remembering something incorrectly, but didn't a similar change on the history of Welsh?

Either way, it seems to make sense to me, but I don't want to do anything with it until I've gotten a second opinion.

[Daedolon]

Could this happen?
k > q
pʰa > pʰa̤ >fa̤
tʰa > tʰa̤ >θa̤
kʰa > kʰa̤ >ka̤

or
ʔa > aˀ or a̰

[k1234567890y]

I have the following diphthongs:
falling: /ai ei oi ui au eu iu/
rising: /ia ie wa we/

(corresponding monophthongs: /a e i o u/

is the combination of the following sound changes plausible?

ai > e:
ei > i:
oi ui > wi
au > o:
eu > u:
iu > i:
Vr > V: / !_V
Cia > Cɛ
Cie > Ci
Cua > Cɔ
Cuo > Cu
e > ɛ
o > ɔ

[StrangerCoug]

Could this happen?
k > q
pʰa > pʰa̤ >fa̤
tʰa > tʰa̤ >θa̤
kʰa > kʰa̤ >ka̤

or
ʔa > aˀ or a̰

I don't expect /k/ /q/ to be unconditional, but I can wrap my head around it when adjacent to back vowels. The rest are plausible to me.

[Zaarin]

Could this happen?
k > q
pʰa > pʰa̤ >fa̤
tʰa > tʰa̤ >θa̤
kʰa > kʰa̤ >ka̤

or
ʔa > aˀ or a̰

I don't expect /k/ /q/ to be unconditional, but I can wrap my head around it when adjacent to back vowels. The rest are plausible to me.

Could be plausible if either /q/ is [k̠] (i.e., if /q/ is actually post-velar/pre-uvular) or something else takes the place of /k/ (cf. k > ʔ, t > k in various Polynesian languages).

[vokzhen]

Could this happen?
k > q
pʰa > pʰa̤ >fa̤
tʰa > tʰa̤ >θa̤
kʰa > kʰa̤ >ka̤

or
ʔa > aˀ or a̰

I don't expect /k/ /q/ to be unconditional, but I can wrap my head around it when adjacent to back vowels. The rest are plausible to me.

I believe I've actually seen k>q as a change in few Austronesian languages, though it'd definitely odd.
And thinking hard about it, I'm not actually sure I've seen aspiration cause breathy vowels. Voiced/breathy consonants can cause breathy vowels, Vh can cause breathy vowels, and breathy vowels can cause aspirated consonants, but I'm not sure I've actually seen aspiration > breathy vowels.

[Pole, the]

Could this happen?
k > q
pʰa > pʰa̤ >fa̤
tʰa > tʰa̤ >θa̤
kʰa > kʰa̤ >ka̤

or
ʔa > aˀ or a̰

I don't expect /k/ /q/ to be unconditional, but I can wrap my head around it when adjacent to back vowels. The rest are plausible to me.

Could be plausible if either /q/ is [k̠] (i.e., if /q/ is actually post-velar/pre-uvular) or something else takes the place of /k/ (cf. k > ʔ, t > k in various Polynesian languages).

Maybe with the contrast /k : kʰ/ being realized as [kˤ ː kʰ] prior to that change?

[Zaarin]

Could this happen?
k > q
pʰa > pʰa̤ >fa̤
tʰa > tʰa̤ >θa̤
kʰa > kʰa̤ >ka̤

or
ʔa > aˀ or a̰

I don't expect /k/ /q/ to be unconditional, but I can wrap my head around it when adjacent to back vowels. The rest are plausible to me.

Could be plausible if either /q/ is [k̠] (i.e., if /q/ is actually post-velar/pre-uvular) or something else takes the place of /k/ (cf. k > ʔ, t > k in various Polynesian languages).

Maybe with the contrast /k : kʰ/ being realized as [kˤ ː kʰ] prior to that change?

Possibly. Also worth noting that there are several languages in the Pacific Northwest that lack plain /k/ despite having a significant number of velar consonants. Cf. Tlingit /kʰ kʷ kʰʷ kʼ kʼʷ x xʷ xʼ xʼʷ/.

[Porphyrogenitos]

Chemakum has a similar velar inventory, but minus even /kʼ/ and /kʰ/. And it doesn't have the other aspirated stops either.

[احمکي ارش-ھجن]

I dunno where else to ask this, but... is there such thing as a language having too many unconditional sound changes? Especially, if there are more of such than conditioned ones?